Functional Imaging of the Fetal Brain

Moheb Costandi, M.Sc.
October 9, 2019

This series of images show developing neural wiring, including the connective tracks needed to send brain signals. Image courtesy of The Developing Human Connectome Project

The development of functional magnetic resonance imaging (fMRI) in the early 1990s revolutionized the field of cognitive neuroscience. The technique detects changes in cerebral blood flow, enabling researchers to infer human brain activity during any given task. Since then, variations of the method have allowed them to visualize the brain in other ways; for example, diffusion tensor imaging (DTI) detects the movement of water molecules, helping researchers map white matter tracts, the large nerve fiber bundles connecting brain regions.

In the past decade or so, technological advances have offered investigators the chance to use these methods in utero (‘in the womb’). The ability to scan the brains of fetuses in the months leading up to birth gives an unprecedented look at the developing brain. Such scanning appears to be safe for both participants, and is already beginning to yield data about brain abnormalities associated with preterm birth. It could eventually provide clues to the origins of conditions such as autism and schizophrenia, which are increasingly being thought of as neurodevelopmental conditions that may arise before birth.

A Moving Target

Performing functional neuroimaging on fetuses is far more challenging than scanning the brains of adults. Researchers must contend not only with spontaneous movements of the fetus, but also with the amniotic fluid, which surrounds the fetus and is in constant motion. The mother’s breathing also causes movements that could potentially blur the scans.

“We image one slice, or a small number of slices, which are cross-sections of the baby’s brain, and we can do that fast enough to freeze the maternal breathing motion and—most of the time—the fetal motion,” says Jo Hajnal, a professor of Imaging Science at King’s College London.

“We acquire a very large number of images covering the brain multiple times over, then put all the data into a computer, and isolate the region of the images that contain the baby’s brain.” This results in an orderly stack of slices, each of which is adjusted and realigned by a computer algorithm, in reference to an atlas of anatomical co-ordinates, to produce a reconstructed image of the fetal brain. (See a video of the process)

Hajnal is a principal investigator on the Developing Human Connectome Project, a collaborative effort with researchers at Imperial College London and Oxford University that aims to create the first dynamic map of human brain connectivity during the second half of pregnancy.

Hajnal and his colleagues and collaborators are focusing on refining their methods, which they are making openly available to other research groups. Earlier this year, they published preliminary data from 140 fetal brains scanned at 38-44 weeks of gestation, together with an automated data processing and analysis method. Ultimately, they hope to combine their scans with behavioral, clinical, and genetic data, so that they and others can “undertake pioneer studies into normal and abnormal development … [in] infants with specific risks that could lead to autism spectrum disorder or cerebral palsy.”

3D reconstruction of the cortical surface and calculated features from aggregated fetal scans at seven months, eight months, and nine months gestation. From top to bottom: white matter surface, cortical surface, inflated surface, distinguishing different structures, how deep the sulci (fissures) are, mean curvature, cortical thickness and T1/T2 myelin maps. Image courtesy of The Developing Human Connectome Project

Predicting Who Will be Born Before Their Time

Using similar methods, researchers elsewhere are beginning to reveal brain abnormalities in fetuses who are subsequently born prematurely. According to the World Health Organization, more than 1 in 10 babies are born preterm, defined as before 37 weeks of gestation. Preterm birth is linked to various neurological complications that can have life-long consequences. Children born preterm are at higher risk of developing autism and attention deficit/hyperactivity disorder (ADHD), for example, and are far more likely to perform poorly at school.

The neurological problems associated with these outcomes were suspected to begin before birth, but this has been very difficult to test. A pioneering 2017 study provided the first direct evidence of brain abnormalities in fetuses that would go on to be born preterm. Moriah Thomason, now at New York University’s School of Medicine, and her colleagues used fMRI to measure brain function in 32 human fetuses, looking specifically at long-range connections. Fourteen of these pregnancies ended in preterm delivery between 24 and 35 weeks of gestation; the remaining 18 reached full term.

The researchers found that functional connectivity in the left hemisphere was markedly reduced in those fetuses that would later be born preterm compared to the others. The reduced connectivity was observed in regions that would eventually contribute to language processing; the strength of the connections was also related to the delivery date: Those with the weakest connectivity were born most prematurely.

More recently, Thomason and her colleagues published a study showing that exposure to lead during pregnancy can alter brain development in the fetus, leading to “alterations in systems that support higher-order cognitive and regulatory functions” which “may account for significant variation in future child cognitive and behavioral outcomes.”

Future work should aim to replicate these early findings in larger, more robust studies. Longitudinal studies involving fetal fMRI followed by behavioral and other tests later in life could strengthen the evidence for a link between brain abnormalities that arise in the womb and subsequent adverse outcomes. One day, they may perhaps also lead to diagnostic tests that identify neurological and psychiatric illness long before birth.

Safety & Ethics

“It would be huge if we could use  fetal fMRI for diagnostic purposes,” says Judy Illes, a professor of neurology and director of Neuroethics Canada at the University of British Columbia in Vancouver. “If it could tell us if a fetus was experiencing pain or stress that may have medical consequences, then there’s enormous merit to it.”

“As long as there appear to be no risks, and women are willing to undergo the procedure, then I’d be hard-pressed to say that it shouldn’t be used for research purposes, too,” says Illes, “but there’d have to be procedures in place to manage unexpected findings.”

According to Hajnal, MRI is a very safe procedure—it has been used thousands of times with young humans with no ill effects—and even the development of increasingly powerful magnets used to scan brains poses no risk. Still, “it’s harder to image large objects with stronger magnets, so imaging fetuses in the next generation of very strong scanners is unlikely to happen.”

Illes points out, however, that we do not know the potential long-term effects of fetal fMRI using the preferred low-strength magnets. “Has anyone done the longitudinal studies? That information is still unknown but should be essential as part of an informed consent process.”